PROJECT SUMMARY HIV latency is a major hurdle to overcome in the efforts to cure AIDS. Latently infected cells do not express viral proteins and escape immune surveillance during HAART treatment. These cells produce infectious viruses upon cessation of the treatment and immune activation. A challenge to achieving an effective HIV anti- latency therapy (HALT) is that these latently infected cells are very difficult to detect. To this end, it is critical to develop a sensitive method to detect latently infected cells in HIV-infected individuals. Although latently infected cells do not express productive viral mRNAs, defective or non-productive viral RNA are expressed and can be detected, which could serve as an excellent biomarker to detect latently infected cells. Highly sensitive next-generation sequencing techniques have revealed that HIV-infected cells express less characterized HIV RNAs including short RNAs, antisense RNAs, host-viral hybrid RNAs and splice variants from cryptic splicing sites, although the role(s) of these RNA species in the viral latency and pathogenesis are largely unknown. However, RT-qPCR analyses of patient cells would often underestimate these aberrant RNAs because of their instability. It is therefore critical to obtain comprehensive catalogues of viral RNA species expressed in latently infected cells using methods that can also detect unstable and short-lived RNA species. My preliminary results of RNA-seq analysis using well-established Jurkat latent models indicate that these cells exclusively express host-viral hybrid RNAs in unstimulated states. Also, reduction of the RNA exosome components by siRNA increased the level of these hybrid RNAs as well as antisense viral RNA. In the proposed study, exact patterns of viral RNA expression (transcriptome) during the establishment of viral latency and its reactivation will be revealed using ex vivo latency models of primary CD4+ cells as well as HIV-infected cells derived from HAART-suppressed patients. In addition, these analyses will determine the proportion of latently infected cells that express these aberrant RNAs, and whether there is a bias in the correlation between HIV RNA expression and provirus integration sites (orientation, genetic environment). Finally, a highly sensitive assay to detect and quantify HIV RNAs expressed specifically in latently infected cells will be developed. These studies will provide important information for developing an efficient therapeutic approach to improve our current HALT regimen.